A redundant fault-tolerant system for industrial grade computer monitoring

Abstract

This invention discloses a redundant fault-tolerant system for industrial-grade computer monitoring. It relates to the field of industrial-grade computer technology and addresses the problem that existing fault-tolerant systems often use manually preset fixed values ​​for test data, failing to define the test range based on historical anomaly scenarios of modules. This invention achieves dynamic matching of fault-tolerant parameters through a percentage-average algorithm, automatically determining the anomaly level of the current module without manual intervention, accurately invoking corresponding fault-tolerant strategies, specifically limiting data request volume or debugging running memory, and quickly pulling abnormal read / write states back to the normal range. This adaptive fault-tolerant mechanism overcomes the shortcomings of traditional fault-tolerant systems with fixed thresholds and one-size-fits-all execution. It customizes fault-tolerant solutions for different modules and different data request scenarios, balancing fault-tolerant efficiency and operational smoothness, ensuring continuous and stable operation of industrial-grade computers under complex conditions, and significantly reducing the probability of unplanned downtime.

Description

Technical Field

[0001] This invention relates to the field of industrial-grade computer technology, specifically to a redundant and fault-tolerant system for industrial-grade computer monitoring. Background Technology

[0002] In core industrial scenarios such as industrial automation, intelligent manufacturing, and large-scale energy storage management, industrial-grade computers serve as the core carriers for data processing, instruction scheduling, and equipment monitoring. Their operational stability and fault tolerance directly determine the continuous operation level of the entire industrial system. Compared to commercial computers, industrial-grade computers need to operate for extended periods in environments with multi-task concurrency, high data throughput, and complex operating conditions. They experience frequent memory read / write operations and diverse data request characteristics. If any working module experiences abnormalities such as read / write performance deviations or response lags, it can easily lead to instruction execution interruptions, data loss, or even systemic shutdowns, causing significant production losses.

[0003] Therefore, building efficient and accurate redundant fault-tolerant systems for industrial-grade computers has become a core technological requirement in the field of industrial control.

[0004] Currently, fault tolerance solutions for industrial-grade computers are mainly divided into two categories: hardware redundancy and software fault tolerance. Hardware redundancy often adopts dual-machine hot standby and multi-module mirror deployment to achieve fault switching through physical backup. However, this method has high hardware costs and cannot intervene in advance for implicit performance deviations of the modules themselves. Software fault tolerance, on the other hand, is based on fixed threshold monitoring. By setting static thresholds for read / write speed and data throughput, it determines whether a module is in an abnormal state and triggers preset fault tolerance strategies, such as forcibly restarting the module or limiting the total number of requests.

[0005] Existing software fault-tolerance technologies have significant limitations in practical applications: First, the anomaly identification lacks precision and specificity. The functional positioning and memory configuration of different working modules inside industrial-grade computers vary significantly, and their normal read and write speeds naturally have inherent ranges. However, traditional solutions use a uniform monitoring threshold, which cannot adapt to the performance differences of each module, nor can it mine hidden deviation features through historical operating data. Often, warnings are only triggered after the fault becomes apparent, missing the best time for intervention. Secondly, redundancy testing is out of touch with actual operating conditions. The test data of existing fault-tolerant systems are mostly fixed values ​​preset by humans, without defining the test range in conjunction with the module's historical abnormal scenarios. As a result, the redundancy characteristics tested cannot truly reflect the module's fault tolerance capability under actual data requests, and the adaptability of the fault tolerance strategy is greatly reduced. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a redundant fault-tolerant system for industrial-grade computer monitoring, which solves the problem that the test data of existing fault-tolerant systems are mostly fixed values ​​preset by humans, and the test range is not defined in conjunction with the historical abnormal scenarios of the modules.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a redundant fault-tolerant system for industrial-grade computer monitoring, comprising: On the historical feature processing end, based on historical cloud data, the read and write capabilities associated with different modules under different memory states are extracted. Simultaneously, deviations from the read and write capabilities associated with individual modules are identified and marked. The specific method is as follows: Identify the working modules existing within industrial-grade computers and extract the associated working data of these modules from historical cloud data; Several historical reading speeds belonging to the same working module are subjected to feature clustering processing. These historical reading speeds are then sorted in ascending order of value to generate a historical reading speed sequence. A data segment is randomly selected from this sequence, and its data range is extracted. This range is the difference between the minimum and maximum historical reading speed. Finally, the total number G of historical reading speeds within the data segment is determined. i Where i represents different data segments, using: M i =Data range ÷G i Lock the density feature M associated with the corresponding data segment. i ; Identify the different density features M associated with different data segments i And from the confirmed density features M i Select the maximum value M i max, and set the maximum value M. i The data segment associated with max is marked as the standard data segment, and the historical reading speed below the standard data segment is recorded as the deviation reading speed. The multiple deviation reading speeds associated with a single working module are recorded as the deviation reading speed set, and the different deviation reading speed sets associated with different working modules are confirmed. The comprehensive feature analysis end identifies different deviations in read speed based on the different sets of read speeds confirmed by different working modules. It then determines the different data request characteristics associated with these deviations and uses the capacity data and read / write length data from these data request characteristics to narrow down the test data range. Specifically: From the set of deviation reading rates associated with a single working module, identify the data request characteristics associated with a single deviation reading rate, and extract the data block capacity R from the data request characteristics. k Where k represents different deviations from the read speed, the read / write lengths of several data segments associated with the data request characteristics are averaged, and the average length C is locked. k ; The capacity R of several data blocks confirmed by a single working module k Perform feature clustering to make several groups of R k The data are sorted in ascending order of value to generate a capacity sequence. Data segments are then randomly selected from this capacity sequence, and the density characteristics of each segment are verified. The maximum value among the verified density characteristics is selected as the standard data segment. The data range associated with the standard data segment is denoted as the capacity range. Using a verification method that confirms identical capacity ranges, the average length C of several sets of values ​​verified by a single working module is determined. k Perform clustering to determine the length range; The capacity and length ranges confirmed by a single working module are recorded as the test data range of the corresponding working module. In the fault-tolerant environment testing phase, based on the confirmed test data range of a single working module and the specific state of the working module's memory, redundancy testing is performed on the working module, and redundancy characteristics are recorded. The specific method is as follows: If the memory inside a single working module is fixed, the minimum value of the capacity range and the minimum value of the length range are confirmed from the test data range. Request data of the corresponding type is generated, and the generated request data is processed for read and write tests through the working module. The single data request volume is gradually adjusted downwards, and the read and write speeds are recorded during the adjustment process until the read and write speeds fall within the set speed range. The speed range is a preset range. Each working module is associated with a corresponding speed range, and the single data request volume is recorded. Then, request data of the corresponding type is generated for the maximum value of the capacity range and the maximum value of the length range. Read and write tests are performed on the working module simultaneously, and the single data request volume after adjustment is recorded. Based on the two recorded single data request volumes, a single data request range is generated, and the redundancy characteristics of the corresponding working module are recorded. If the memory inside a single working module is adjustable, two sets of corresponding request data are generated based on the test data range associated with the corresponding working module, and read / write test processing is performed through this working module. During the test processing, the running memory of this working module is gradually increased until the read / write speed associated with this working module falls within the set read speed range, and the two sets of running memory that are increased are recorded as the redundancy characteristics of the corresponding working module. The monitoring center monitors the read and write speeds of different working modules within an industrial-grade computer in real time. The fault-tolerant feature execution end confirms the read / write speed of the corresponding working module against the deviation read speed set. If the read / write speed falls within the numerical range associated with the deviation read speed set, the fault-tolerant feature is confirmed and executed, putting the working module into a normal read / write state. Specifically: Identify the data request characteristics associated with the current working module, and locate the data block capacity within the capacity range from these characteristics: define the capacity range as [R]. k min,R k [max], mark the current database capacity as RB, and use: (RB-R) k min) ÷ (R) k max-R k min) = ZB1, confirm the first percentage value; Next, lock the average length of the read / write operations of several data segments, and record the position of the average length within the length range: Define the length range as [C]. k min,C k [max], denoted as CJ, the length of the mean value in this case is defined as: (CJ-C k min) ÷ (C k max-C k min) = ZB2, confirm the first percentage value; The first proportion value ZB1 and the second proportion value ZB2 are averaged to confirm the average proportion. Based on the redundancy features recorded by the corresponding working module, a fault-tolerant feature is selected from the numerical range associated with the redundancy feature, and the redundancy feature is denoted as [RY]. k min,RY k [max], label the fault-tolerant feature as ZY, and the selected fault-tolerant feature satisfies: (ZY-RY) k min) ÷ (RY k max-RY k min) = mean percentage, and execute the selected fault tolerance feature.

[0008] Preferably, the working data includes historical read speed and data request characteristics, including the overall capacity of the data block and the read / write length of a single data segment.

[0009] This invention provides a redundant fault-tolerant system for industrial-grade computer monitoring. Compared with existing technologies, it has the following advantages: Based on the data request characteristics associated with deviations in read speed, a secondary clustering algorithm is used to accurately define the test data range for each working module, taking into account both data block capacity and single read / write length to ensure a high degree of match between the test scenario and the actual operating conditions of industrial computers. For modules with fixed memory and adjustable memory, a differentiated redundancy testing strategy is adopted in the fault-tolerant environment test: the fixed memory module locks the safe operating threshold by adjusting the request volume in a gradient manner, while the adjustable memory module adapts to normal read / write standards by dynamically increasing memory usage. The testing process eliminates the need for blind trial and error, ensuring the authenticity of redundant feature collection while significantly improving testing efficiency and reducing test losses. By combining real-time monitoring data with a redundant feature library, a proportional-average algorithm is used to dynamically match fault tolerance parameters. This automatically determines the anomaly level of the current module without manual intervention, accurately invoking the corresponding fault tolerance strategy to specifically limit data request volume or debug runtime memory, quickly bringing abnormal read / write states back to normal ranges. This adaptive fault tolerance mechanism overcomes the shortcomings of traditional fault tolerance systems with fixed thresholds and one-size-fits-all execution. It customizes fault tolerance solutions for different modules and data request scenarios, balancing fault tolerance efficiency and operational smoothness, ensuring continuous and stable operation of industrial-grade computers under complex conditions, and significantly reducing the probability of unplanned downtime. Attached Figure Description

[0010] Figure 1 This is a schematic diagram of the principle framework of the present invention. Detailed Implementation

[0011] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0012] First Embodiment Please see Figure 1 This application provides a redundant fault-tolerant system for industrial-grade computer monitoring, including a historical feature processing terminal, a comprehensive feature analysis terminal, a fault-tolerant environment testing terminal, a fault-tolerant feature execution terminal, and a monitoring center. The historical feature processing terminal, the comprehensive feature analysis terminal, the fault-tolerant environment testing terminal, and the fault-tolerant feature execution terminal are electrically connected sequentially from the output node to the input node, and the monitoring center is electrically connected to the input node of the fault-tolerant feature execution terminal. Among them, the historical feature processing end extracts the read and write capabilities associated with different modules in different memory states based on the different historical cloud data associated with different modules in the industrial-grade computer. Simultaneously, it locks and marks the set of deviations from the read and write capabilities associated with a single module. Specifically, in the usage state of certain modules, based on historical cloud data, it identifies the specific parameters of historical read speed deviation, thereby identifying some data features that are more difficult for the corresponding module to process. The specific method for locking the reading speed deviation is as follows: Identify the working modules existing in the industrial-grade computer and extract the working data associated with the working modules from historical cloud data. The working data includes historical read speed and data request characteristics. The data request characteristics include the overall capacity of the data block and the read / write length of a single data segment. Several historical reading speeds belonging to the same working module are subjected to feature clustering processing. These historical reading speeds are then sorted in ascending order of value to generate a historical reading speed sequence. A data segment is randomly selected from this sequence, and its data range is extracted. This range is the difference between the minimum and maximum historical reading speed. Finally, the total number G of historical reading speeds within the data segment is determined. i Where i represents different data segments, using: M i =Data range ÷G i Lock the density feature M associated with the corresponding data segment. i ; Identify the different density features M associated with different data segments i And from the confirmed density features M i Select the maximum value M i max, and set the maximum value M. i The data segment associated with max is marked as the standard data segment, and the historical reading speed below the standard data segment is recorded as the deviation reading speed. The multiple deviation reading speeds associated with a single working module are recorded as the deviation reading speed set, and the different deviation reading speed sets associated with different working modules are confirmed. Specifically, within a historical reading speed sequence, there are several different data segments, and each data segment is associated with different density features. The data segment whose density feature is at its maximum value is the most clustered data segment. Such data segments represent that the corresponding working module is in normal operation. Based on this normal state, the deviation reading speed can be quickly and effectively identified, and the associated deviation reading speed set can be confirmed, which facilitates subsequent feature confirmation.

[0013] Among them, the comprehensive feature analysis end, based on the different deviation read speed sets identified by different working modules, identifies different data request characteristics associated with different deviation read speeds, and locks the test data range from the capacity data and read / write length data in the data request characteristics. The locked test data range is then transmitted to the fault-tolerant environment test end. The specific method for locking the test data range is as follows: From the set of deviation reading rates associated with a single working module, identify the data request characteristics associated with a single deviation reading rate, and extract the data block capacity R from the data request characteristics. k Where k represents different deviations from the read speed, the read / write lengths of several data segments associated with the data request characteristics are averaged, and the average length C is locked. k ; The capacity R of several data blocks confirmed by a single working module k Perform feature clustering to make several groups of R kThe data are sorted in ascending order of value to generate a capacity sequence. Data segments are then randomly selected from this capacity sequence, and the density characteristics of each segment are verified. The maximum value among the verified density characteristics is selected as the standard data segment. The data range associated with the standard data segment is denoted as the capacity range. Using a verification method that confirms identical capacity ranges, the average length C of several sets of values ​​verified by a single working module is determined. k Perform clustering to determine the length range; The capacity and length ranges confirmed by a single working module are recorded as the test data range of the corresponding working module. Specifically, each working module has a single set of deviation reading speeds. Each deviation reading speed in the set is associated with a set of data request characteristics. Each data request characteristic is associated with a data request process, and each data request process is associated with data request data. The data request data is the capacity of the associated data block and the read / write length of a single data segment. This effectively confirms the corresponding length range and capacity range, which facilitates the subsequent confirmation of the corresponding test data range and makes it easier to perform redundancy testing on the working module.

[0014] Second Embodiment In this embodiment, compared to the above embodiments, the main focus is on the redundancy testing process of the working modules. Among them, the fault-tolerant environment test terminal performs redundancy testing on the working module based on the test data range confirmed by a single working module and the specific state of the working module's memory, and records the redundancy characteristics. The specific method for handling redundancy testing is as follows: If the memory within a single working module is fixed (i.e., the memory cannot be debugged), confirm the minimum value of the capacity range and the minimum value of the length range from the test data range, generate the corresponding type of request data, and process the generated request data through the working module for read and write tests. Gradually debug the single data request volume downwards, and record the read and write speeds during the debugging process until the read and write speeds fall within the set read speed range. The read speed range is a preset range. Each working module is associated with a corresponding read speed range, and the debugged single data request volume is recorded. Then, generate the request data of the type corresponding to the maximum value of the capacity range and the maximum value of the length range, and synchronously perform read and write tests on the working module, and record the single data request volume after debugging. Based on the two recorded single data request volumes, generate the single data request range and record the redundancy characteristics of the corresponding working module. If the memory within a single working module is adjustable, two sets of corresponding request data are generated based on the test data range associated with the working module. These requests are then processed through the working module for read / write testing. During the testing process, the running memory of this working module is gradually increased until the read / write speed associated with this working module falls within the set read speed range. The two sets of increased running memory are recorded as the redundancy characteristics of the corresponding working module. If the read / write speed cannot reach the set read speed range during the testing process, a memory request signal is generated and output. External personnel can then use this memory request signal to debug the adjustable memory of this working module, causing the corresponding adjustable memory to be increased.

[0015] The monitoring center monitors the read and write speeds of different working modules within the industrial-grade computer in real time and transmits the monitored read and write speeds to the fault-tolerant execution terminal. Among them, the fault tolerance feature execution end confirms the read and write speed of the corresponding working module and the deviation read speed set. If the read and write speed is within the numerical range associated with the deviation read speed set, the fault tolerance feature is confirmed and executed to put the working module into a normal read and write state. The specific method for confirming fault-tolerant features is as follows: Identify the data request characteristics associated with the current working module, and locate the data block capacity within the capacity range from these characteristics: define the capacity range as [R]. k min,R k [max], mark the current database capacity as RB, and use: (RB-R) k min) ÷ (R) k max-R k min) = ZB1, confirm the first percentage value; Next, lock the average length of the read / write operations of several data segments, and record the position of the average length within the length range: Define the length range as [C]. k min,C k [max], denoted as CJ, the length of the mean value in this case is defined as: (CJ-C k min) ÷ (C k max-C k min) = ZB2, confirm the first percentage value; The first proportion value ZB1 and the second proportion value ZB2 are averaged to confirm the average proportion. Based on the redundancy features recorded by the corresponding working module, a fault-tolerant feature is selected from the numerical range associated with the redundancy feature, and the redundancy feature is denoted as [RY]. k min,RY k[max], label the fault-tolerant feature as ZY, and the selected fault-tolerant feature satisfies: (ZY-RY) k min) ÷ (RY k max-RY k min) = mean percentage, and apply the selected fault tolerance feature; Specifically, when the read / write speed of the corresponding working module deviates from the expected state, the corresponding execution parameters need to be adjusted. This could involve limiting the number of requests per transaction or adjusting the corresponding execution memory. Based on the confirmed data, the working status of the working module is adjusted to ensure that the industrial-grade computer has sufficient redundancy and fault tolerance capabilities.

[0016] Some of the data in the above formulas are numerical calculations with dimensions removed, and the contents not described in detail in this specification are all prior art known to those skilled in the art.

[0017] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A redundant fault-tolerant system for industrial-grade computer monitoring, characterized in that, include: On the historical feature processing end, based on historical cloud data, the read and write capabilities associated with different modules under different memory states are extracted, and the deviation set of read speed is locked and marked from the read and write capabilities associated with a single module. The comprehensive feature analysis end identifies different data request features associated with different deviations in reading speed based on the different sets of reading speed deviations confirmed by different working modules, and locks out the test data range from the capacity data and read / write length data in the data request features. The fault-tolerant environment testing end performs redundancy testing on the working modules based on the test data range confirmed by a single working module and the specific state of the working module's memory, and records the redundancy characteristics. The monitoring center monitors the read and write speeds of different working modules within an industrial-grade computer in real time. The fault-tolerant feature execution end confirms the read / write speed of the corresponding working module and the deviation from the read speed set. If the read / write speed is within the numerical range associated with the deviation from the read speed set, the fault-tolerant feature is confirmed and executed, so that the working module is in normal read / write state.

2. The redundant fault-tolerant system for industrial-grade computer monitoring according to claim 1, characterized in that, The historical feature processing terminal uses the following specific method to lock deviations in reading speed: Identify the working modules existing within industrial-grade computers and extract the associated working data of these modules from historical cloud data; Several historical reading speeds belonging to the same working module are subjected to feature clustering processing. These historical reading speeds are then sorted in ascending order of value to generate a historical reading speed sequence. A data segment is randomly selected from this sequence, and its data range is extracted. This range is the difference between the minimum and maximum historical reading speed. Finally, the total number G of historical reading speeds within the data segment is determined. i Where i represents different data segments, using: M i =Data range ÷G i Lock the density feature M associated with the corresponding data segment. i ; Identify the different density features M associated with different data segments i And from the confirmed density features M i Select the maximum value M i max, and set the maximum value M. i The data segment associated with max is marked as the standard data segment, and the historical reading speed below the standard data segment is recorded as the deviation reading speed. The multiple deviation reading speeds associated with a single working module are recorded as the deviation reading speed set, and the different deviation reading speed sets associated with different working modules are confirmed.

3. The redundant fault-tolerant system for industrial-grade computer monitoring according to claim 1, characterized in that, The working data includes historical read speed and data request characteristics, including the overall capacity of the data block and the read / write length of a single data segment.

4. The redundant fault-tolerant system for industrial-grade computer monitoring according to claim 1, characterized in that, The specific method by which the comprehensive feature analysis terminal locks the range of test data is as follows: From the set of deviation reading rates associated with a single working module, identify the data request characteristics associated with a single deviation reading rate, and extract the data block capacity R from the data request characteristics. k Where k represents different deviations from the read speed, the read / write lengths of several data segments associated with the data request characteristics are averaged, and the average length C is locked. k ; The capacity R of several data blocks confirmed by a single working module k Perform feature clustering to make several groups of R k The data are sorted in ascending order of value to generate a capacity sequence. Data segments are then randomly selected from this capacity sequence, and the density characteristics of each segment are verified. The maximum value among the verified density characteristics is selected as the standard data segment. The data range associated with the standard data segment is denoted as the capacity range. Using a verification method that confirms identical capacity ranges, the average length C of several sets of values ​​verified by a single working module is determined. k Perform clustering to determine the length range; The capacity and length ranges confirmed by a single working module are recorded as the test data range of the corresponding working module.

5. The redundant fault-tolerant system for industrial-grade computer monitoring according to claim 1, characterized in that, The specific method for recording redundancy features in the fault-tolerant environment testing terminal is as follows: If the memory within a single working module is fixed, determine the minimum value of the capacity range and the minimum value of the length range from the test data range, generate corresponding request data, and process the generated request data through the working module for read and write tests. Gradually debug the single data request volume downwards, and record the read and write speeds during the debugging process until the read and write speeds fall within the set read speed range. The read speed range is a preset range. Each working module is associated with a corresponding read speed range, and the debugged single data request volume is recorded. Then, generate request data of the corresponding type with the maximum value of the capacity range and the maximum value of the length range, and synchronously perform read and write tests on the working module, and record the debugged single data request volume. Based on the two recorded single data request volumes, generate the single data request range and record the redundancy characteristics of the corresponding working module.

6. The redundant fault-tolerant system for industrial-grade computer monitoring according to claim 5, characterized in that, The specific methods for recording redundancy features in the fault-tolerant environment testing terminal also include: If the memory inside a single working module is adjustable, two sets of corresponding request data are generated based on the test data range associated with the corresponding working module. Read and write tests are then performed through this working module. During the test process, the running memory of this working module is gradually increased until the read and write speed associated with this working module falls within the set read speed range. The two sets of running memory that have been increased are recorded as the redundancy characteristics of the corresponding working module.

7. The redundant fault-tolerant system for industrial-grade computer monitoring according to claim 1, characterized in that, The specific method by which the fault-tolerant feature execution terminal confirms the fault-tolerant feature is as follows: Identify the data request characteristics associated with the current working module, and locate the data block capacity within the capacity range from these characteristics: define the capacity range as [R]. k min,R k [max], mark the current database capacity as RB, and use: (RB-R) k min) ÷ (R) k max-R k min) = ZB1, confirm the first percentage value; Next, lock the average length of the read / write operations of several data segments, and record the position of the average length within the length range: Define the length range as [C]. k min,C k [max], denoted as CJ, the length of the mean value in this case is defined as: (CJ-C k min) ÷ (C k max-C k min) = ZB2, confirm the first percentage value.

8. The redundant fault-tolerant system for industrial-grade computer monitoring according to claim 7, characterized in that, The specific methods for confirming the fault tolerance feature at the fault tolerance feature execution end also include: The first proportion value ZB1 and the second proportion value ZB2 are averaged to confirm the average proportion. Based on the redundancy features recorded by the corresponding working module, a fault-tolerant feature is selected from the numerical range associated with the redundancy feature, and the redundancy feature is denoted as [RY]. k min,RY k [max], label the fault-tolerant feature as ZY, and the selected fault-tolerant feature satisfies: (ZY-RY) k min) ÷ (RY k max-RY k min) = mean percentage, and execute the selected fault tolerance feature.

Images